Jump to content

Why is AAA development so insane? And is there an alternative?


Recommended Posts

33 minutes ago, Popo said:


I broadly understand what you’re saying about agile, and it totally makes sense, but I’m not sure I understand why it precludes making decisions about the scope of the work at the beginning? Your example above illustrates this - if a balcony wasn’t originally in scope it’s very easy to identify why a huge time and resource investment would be needed to redesign and build from scratch. 

 

I’m a project manager so I have a natural bias against scope creep and deviations from spec, but both are completely reasonable so long as everyone involved understands that changing the scope later implies delay and increased cost. 

 

The waterfall way, where you ask the client to make all the decisions up front sounds great in principle. You explain to the client that you absolutely need to know everything EVERYTHING that needs to go into this project, because if they change their mind later, it's REALLY expensive.

 

So the client sits down and stresses the details. Like ALL the details. And they come back with this huge list of things. And it takes years and years to build it to spec, and at the end of it half of the things the client said they wanted end up being redundant because you/they weren't able to prototype and discover what was valuable and what was not.

 

Agile helps here by building what the client says is important first, and iterating on that before moving onto the next thing. You can still have a list of ambitions ( backlog items in Scrum terminology), but this list is dynamic depending on what the client decides is valuable after evaluating the current state of the product.  The reality is that the lower priority items are often not required at all, and via prototyping the client is able to realise this and not waste time developing them.

Link to post
Share on other sites
11 minutes ago, The Bag said:


Bane of my fucking existence, I currently work with 2 of the best UI artists I’ve ever worked with but duck me what they want to do regularly pushed unity’s ui to breaking point.

If you want to get rich in this business, write a great UI framework!

Link to post
Share on other sites
3 hours ago, Broker said:


Fable is really exciting because the team have extensive experience of triple A but mostly they’ve made racing games. I’m really interested to see what they do, although I hope it’s not too twee and ARRY POTTA.

 

Given they've just taken on Anna Megill (narrative lead for Control, writer on Dishonored: Death of the Outsider) as their lead writer I think it's got a good chance to have excellent writing.

 

(hopefully they've not done the traditional "design and structure your game, then hire a writer to fill in the gaps" approach which inevitably stymies the narrative potential)

Link to post
Share on other sites

Larian's Divinity 2 and Baldur's Gate III show us how AAA can be done in 2020. 

 

That we accept the shitshows trainwreck releases of Take 2, EA, Activision, Ubisoft and CDPR is all on us. 

 

#votewithyourwallet is the only right way forward. Don't pre-order, don't buy on day 1. 

Link to post
Share on other sites
14 minutes ago, Wiper said:

 

Given they've just taken on Anna Megill (narrative lead for Control, writer on Dishonored: Death of the Outsider) as their lead writer I think it's got a good chance to have excellent writing.

 

(hopefully they've not done the traditional "design and structure your game, then hire a writer to fill in the gaps" approach which inevitably stymies the narrative potential)


Also Playground have built a second team to do Fable, as I understand it they gave their existing team the choice on whether to go onto it or not.  What I hope they get is Dene’s major influence for Fable’s setting is British folk tales which are by and large pretty fucked up and nasty.

Link to post
Share on other sites
2 hours ago, Uncle Mike said:

How do you know someone works in Agile?

 

Don't worry, they'll tell you. Etc.

 

In fairness, the thread starter did ask why AAA studios use Agile instead of Waterfall. He just didn't know that was what he was asking :P

Link to post
Share on other sites
46 minutes ago, Wiper said:

 

Given they've just taken on Anna Megill (narrative lead for Control, writer on Dishonored: Death of the Outsider) as their lead writer I think it's got a good chance to have excellent writing.

 

(hopefully they've not done the traditional "design and structure your game, then hire a writer to fill in the gaps" approach which inevitably stymies the narrative potential)


I’m writing my dissertation on the potential benefits of incorporating narrative frameworks into the development process and using than methodology to create ludonarrative resonance. There’s fuck all sources that have any examples of the development process including narrative structure in their planning that aren’t David Cage style story games, but I read a really interesting paper about aligning narrative goals with player goals in order to scaffold the players actions in a narrative context.

Link to post
Share on other sites

From my experience within the games industry its because as soon as a project is greenlit its all hands on deck to get it made. It's so backwards, so much time is wasted on these huge scopes and things being made before being ripped out. Whole levels or systems that are nearly finished just removed. Thats years wasted that could be spent making the core game great.

 

Games should be made with a tight modular focus. Got a racing game for example. Just make 1 track and 1 car and make it perfect. Then just add on extra tracks and cars as time permits. Currently in this scenerio you go in making 1000 cars and 100 tracks and start work on all of them at once and before long you have to cut it down to 100 cars and 10 tracks with all this work on the cutting room floor.

 

It's the same with gameplay mechanics, i remember one game i work on having...lets say 'swings' to avoid me giving away what game it was. Someone up to decided there needed to be a swing mechanic in the game, now off you go and put it in. Even just the basic idea didn't make much sense and didn't sound fun and whilst most the dev team were against it just from the sound of it they had to put it in. Now if they could have just prototyped it for a week they could have either gone back and shown how its no fun or found a way to make it fun. Instead if got put in the game as is and isn't fun and had no love or attention put into it. It basically just ticked a box somewhere.

 

And thats the main thing with game development, most games that are great, are great because the devs either had a prototype or had a small idea someone worked on in their spare time that showed something fun happening. Think of the original GTA, that was just the devs messing around at the start and landing on fun gameplay. They didn't set out to make GTA as it was and if they had it'd probably have been dull, the fun came from what they put in because it was fun to them.

 

You certainly see this with japanese games vs the west. Take Shenmue 3, when it first was shown the graphics were terrible, or take Mario 3D World and the graphics were pretty average. Thats because those devs put in what was needed to get up and running and then later on added on top to make everything look much nicer. Mario 3D world in particular looks great as a finished game. Where as western devs tend to go the other way, putting their best stuff in first, which takes time and then having to rip out stuff because they ran out of time. Basically a lot of this boils down to poor product planning, mismanagement and most devs just having to react to making stuff on the fly rather than a concrete plan of action. The movie analogy is perfect for this, games should follow this with storyboards and showing how ideas work so when you go and make the product you are making only the best content, not the cut content. Films like Toy Story don't really have cut content because all that stuff was cut at the storyboard stage.

Link to post
Share on other sites
2 hours ago, Uncle Mike said:

How do you know someone works in Agile?

 

Don't worry, they'll tell you. Etc.


Yeah, for developers creating new product in areas they know well, or can integrate domain experts into the teams, it's great.

 

Waterfall works for some teams and projects - basically, depends on culture.

Link to post
Share on other sites
1 hour ago, Plissken said:

I'm somewhat reluctant to write given that it might seem like flamebait so I'm going to choose my words very carefully but this thread seems to be pinning various aspects of the blame on studios, producers, developers, marketing etc etc and missing out one very vital component.

 

Gamers.

 

I've been reading the Cyberpunk thread with interest and it seems to replicate lots of previous threads and comments in gaming generally - a lot of the most vocal gamers are quite happy to take the shinies and prioritise FPS, 4K and the latest visual tech over an actual solid game playing experience.  They don't care that the game is full of bugs and crashes as long as it pushes the technology in their particular box of choice past someone elses box of choice.  As long as the disembodied hands and cock are floating in midair at an acceptable framerate in fully ray-traced 4K they aren't that fussed.  Such things as not making it crash twice an hour can be patched in later.

 

And in this case, I'm not sure I can blame the studios.  I mean, can you imagine the fuss is the issues with Cyberpunk were the other way around?  If it was a comparatively fully formed, mostly bug free, fully laden experience that didn't quite look like an amazing Hollywood movie?  If CDPR said "Yes, it doesn't match the visual quality of Blade Runner 2049, but we concentrated on making the gameplay solid and will patch in the various HDR and 4K things later"?  Various PS5, XBox Series X and PC owners across the internet would be throwing a shitfit you could see from the Moon.

 

Earlier in the thread, people were discussing the BOTW stuff and it's all well and good but the demand from the Entitled Gamer Community is so loud and so vociferous that the people who decide where the money goes are prioritising the wrong thing because of it.  Every time I see a Digital Foundry video my heart sinks because it is about performance and numbers and not whether the bloody thing works.  (There is a half-written blog post I've got somewhere about how the concept of metrics is ruining everything - everything has to be modelled and scored and ranked and this has to be declared better than that and it's sucking all the fun out of life.)  Why is OK to patch in fixes for gameplay six months or a year later and not a visual or audio upgrade?  Why is it OK with games to force developers into a six month crunch to produce a game that is so focused on looking and sounding good that it simply doesn't work properly?  Because you can guarantee that that crunch was focused on optimising the visuals so that it could score the required fps metrics.  We hear so much of development time and cost is focused on the artwork and not the way that it plays and that feels wrong.

 

I'm not excusing the management CDPR in any way here.  What I am saying is that they knew that what they were trying to do wouldn't work properly on a bog standard PS4 or XBox One.  What am I saying is why did they feel compelled to hide it?  Why did they feel that it was better to release something utterly broken than cut down on the shinies and the scope.  And I think that even the fact that said PS4 and XBox One consoles are referred to as "poverty spec" has something to do with it.

 

It feels like the cart is put before the horse so many times - even when you have examples, mainly from Nintendo, where the opposite has produced some utter classics.

I feel you, but at the same time I think it's a bit chicken-and-egg though, right?

 

Cos I mean, people's expectations don't set themselves. 

Link to post
Share on other sites
1 hour ago, Plissken said:

I'm not excusing the management CDPR in any way here.  What I am saying is that they knew that what they were trying to do wouldn't work properly on a bog standard PS4 or XBox One.  What am I saying is why did they feel compelled to hide it?  Why did they feel that it was better to release something utterly broken than cut down on the shinies and the scope.  And I think that even the fact that said PS4 and XBox One consoles are referred to as "poverty spec" has something to do with it.

 

The game shouldn't been release on current-gen consoles. They have a solid backup with Stadia, GeForce Now etc... even Switch allows streaming of AAA games. 

Link to post
Share on other sites
33 minutes ago, robdood said:

I feel you, but at the same time I think it's a bit chicken-and-egg though, right?

 

Cos I mean, people's expectations don't set themselves. 

I was going to say exactly this. And for all the industry is pushing the limits on just how far graphical fidelity can be taken, and for what we assume is a vociferous community demanding that games have these features, many games that have exploded in popularity and utterly defined the zeitgeist of the times have defiantly not had these features. Minecraft, PUBG & Fortnite, Fall Guys, Among Us etc have come from nowhere to absolutely dominate the market, and you can certainly argue that they embody that spirit of 'is it fun to play?' first and foremost.

 

In a very specific part of the market, there is a desire to push tech that is felt from both studios and their audience (I do feel like Cyberpunk would not quite have gained the buzz it did had it not been attempting to strive for cutting edge presentation), it is far from the be all and all of what can capture the market. While not being juggernauts of the same calibre, there are also indie titles that have sold gargantuan numbers in comparison to their development budget (I am immediately thinking of Stardew Valley here), and these titles often flourish by appealing to a customer that the larger AAA productions have left unattended for too long. In that instance I recall Amnesia The Dark Descent pulling in incredible sales figures at a time when most if not all major publishers had completely abandoned the horror genre.

Link to post
Share on other sites
2 hours ago, Plissken said:

I'm not excusing the management CDPR in any way here.  What I am saying is that they knew that what they were trying to do wouldn't work properly on a bog standard PS4 or XBox One.  What am I saying is why did they feel compelled to hide it?  Why did they feel that it was better to release something utterly broken than cut down on the shinies and the scope.  And I think that even the fact that said PS4 and XBox One consoles are referred to as "poverty spec" has something to do with it.

 

 

 

They probably actually did think they could do it, it's just the goal posts moved. When this has a what, 2018 release date? Even a 2019 release date they'd have focused on those consoles. But as it slipped further and further they then had to take their limited resources (employees) and stretch them to Stadia, Xbox Series and PS5. Maybe even the new PC graphics cards too. Those consoles made sense at the time and clearly they were only still released now because they'd probably pumped in years of time and money into them. To go from 150 million console users to what, 2 million of the new consoles is not a smart move.

 

Look at GTA 5, that released weeks before the new consoles but they didn't say anything about those consoles getting GTA 5 for like a year. Maybe CDPR should have just focused on the base consoles and totally ignored Xboxs Series and PS5. Despite both being in BC mode, they still have done some work to those versions to make sure they work better and will continue to do so.

Link to post
Share on other sites
1 hour ago, robdood said:

Agile software dev is great in principle, but in my experience can only truly work when a project has unlimited time / budget.  

 

I’ve worked on an Agile project that had both, and after eight years it switched studios twice, creative leads three times, engines once, became rejigged to fit a major entertainment IP, and last I heard they’d scrapped it all and started over again.
 

Agile’s biggest problem for AAA games is that you need to deliver to customers regularly for it to be effective, and most game projects outside of the mobile space don’t think about delivering anything to a customer until it’s essentially functionally and feature complete. Even live service games- how many MMOs emerge from six years of Area 51 secrecy and it turns out players hate 80% of the features? I’m not sure it is the solution any more.

Link to post
Share on other sites
5 hours ago, Ketchup said:

Someone up to decided there needed to be a swing mechanic in the game, now off you go and put it in. 

 

How did that someone arrive at the conclusion that there needs to be a swing mechanic? Did they do market research and found out that players like it?

Link to post
Share on other sites
21 minutes ago, alex3d said:

How did that someone arrive at the conclusion that there needs to be a swing mechanic? Did they do market research and found out that players like it?

Tbh I don’t know the full details of how they came to that conclusion but I highly doubt that was the case. In my experience especially with producers or publishers it’s generally

someone not directly part of the team wanting to put their mark on a product and either coming up with a stupid idea that doesn’t fit or seeing something or liking an idea and asking for it to be included with little thought put into it except to be able to go back to their higher ups and feedback how they improved the game with their added new feature.

 

Theres even evidence of this with rockstar or even peter molyneux where someone high up has an idea they think will be great (cars running out of petrol in gta, trees growing in real time in fable) and whilst it may sound great ultimately as ideas they are terrible with even a few minutes thought.

Link to post
Share on other sites
20 hours ago, MNFRMTMRRW said:

The game shouldn't been release on current-gen consoles. They have a solid backup with Stadia, GeForce Now etc... even Switch allows streaming of AAA games. 

 

When was the game announced and when was it developed?  What is the total userbase of those now against the PS4 and XBox One?  Why are companies aiming for technology that only exists three years after starting development?

 

GTA V gets compared a lot - Rockstar got it right.  Current consoles even though they were coming to the end of the generation and the next generation a year later to keep the money train rumbling.

Link to post
Share on other sites

It's because creating AAA games is hard. They're multi-million pound feats of cutting-edge engineering and creativity involving hundreds, if not thousands, of staff.

 

Given that large IT/software projects are notorious for having major issues, it's honestly a miracle that most AAA games turn out so well.

Link to post
Share on other sites
35 minutes ago, Plissken said:

 

When was the game announced and when was it developed?  What is the total userbase of those now against the PS4 and XBox One?  Why are companies aiming for technology that only exists three years after starting development?

 

GTA V gets compared a lot - Rockstar got it right.  Current consoles even though they were coming to the end of the generation and the next generation a year later to keep the money train rumbling.

From snippets that have come out it sounds like the heavy lifting phases of development of cyberpunk really only started in 2019. Just the 7 years after announcement where they farted about in pre production and alpha stages. 

Link to post
Share on other sites

There was a thread from 2018 about how it feels to spend a year working on a game knowing that it's not very good and a number of replies said that games come together in the last few weeks of development, even games like GTA don't work until it all starts to come together. Perhaps that's an issue right there, that the developers simply don't have a clue for 95% of the dev time whether what they have works or not? Then 2 weeks before release it al comes together and everyone realises , "Oh my God, its broken!" for the first time.

Link to post
Share on other sites
18 hours ago, dumpster said:

There was a thread from 2018 about how it feels to spend a year working on a game knowing that it's not very good and a number of replies said that games come together in the last few weeks of development, even games like GTA don't work until it all starts to come together. Perhaps that's an issue right there, that the developers simply don't have a clue for 95% of the dev time whether what they have works or not? Then 2 weeks before release it al comes together and everyone realises , "Oh my God, its broken!" for the first time.

You do need to have some faith that it's all going to come together. You can only really work on your bits and help on the bits they are connected to. You can't go up to someone you're not connected to and say "when is your bit going to be less shit? want some opinions?" ...you have to trust everyone else around you.

(but you can get into some random fun cross-office chat, as long as they initiate it)

 

I don't think it's quite as bad as it not being fun until right at the end. It can just feel like that if you're working on it. You don't get to play the game. You end up having to race through sections to get to the bits you need to work on. It's hard to get the atmosphere and feel of it when you're trying to speed run just to hit the crash you're looking for. 

 

On 19/12/2020 at 14:08, Plissken said:

I've been reading the Cyberpunk thread with interest and it seems to replicate lots of previous threads and comments in gaming generally - a lot of the most vocal gamers are quite happy to take the shinies and prioritise FPS, 4K and the latest visual tech over an actual solid game playing experience.  They don't care that the game is full of bugs and crashes as long as it pushes the technology in their particular box of choice past someone elses box of choice.  As long as the disembodied hands and cock are floating in midair at an acceptable framerate in fully ray-traced 4K they aren't that fussed.  Such things as not making it crash twice an hour can be patched in later.

In fairness, the core solid gameplay experience looks like it was catered for too in Cyberpunk. From the people who can play it, they enjoyed it. I'm actually more interested in playing now (now as in next year) than I was before release. 

 

Work is pretty compartmentalised too. The people making it shiny have little to do with the game play, and visa versa. This isn't a bad thing, it's just ray-tracing a dick doesn't need any gameplay, and gameplay can get on happily with standard dick tech. 

 

But prioritisation does mean that getting it to look AND FEEL nice might be more important in the early days than having it not crash after 20-30 mins of play (less time: needs fixing now, probably something obvious rather than rot over time). That early on, you can end up wasting time fixing stuff that may not end up in the final game and if that bug doesn't crash, the next completely different one will moments later anyway. 

 

There should be a point where it flips the other way. Stop adding stuff, make sure it works. And that's where they went wrong :/ 

Link to post
Share on other sites
On 18/12/2020 at 14:58, Marlowe said:

 

Was my first thought as well. :lol:

 

I dread starting any kind of 'AAA' game now, as the intros are invariably padded out with boring cinematic stuff and clumsy tutorials that teach you the mechanics in the least natural way possible, despite modern games all basically having the same control schemes anyway. Super Mario 64 was a lot of people's first experience of a 3D game, or at least of controlling a game in that way, yet Nintendo had enough confidence in players to simple drop you in the Castle Gardens and let you discover the mechanics yourself. Same with Liberty Island in Deus Ex. I don't buy the idea that modern games are vastly more complicated than older games (they're not) or that players now are less willing to learn through experimentation and so need more hand holding.

I'm sorry to be this person, but I've just started Mario 64 on the all stars collection, and it begins with a fairly lengthy unskippable cinematic/flyby, then some unskippable tutorial text, then some further unskippable tutorial text on the way in to the castle. There's a lot about it that's not good.

Link to post
Share on other sites
1 hour ago, jonny_rat said:

I'm sorry to be this person, but I've just started Mario 64 on the all stars collection, and it begins with a fairly lengthy unskippable cinematic/flyby, then some unskippable tutorial text, then some further unskippable tutorial text on the way in to the castle. There's a lot about it that's not good.

 

ah come on, at the time it was a new controller and they were showing off their new 3D world...

Link to post
Share on other sites
1 hour ago, jonny_rat said:

I'm sorry to be this person, but I've just started Mario 64 on the all stars collection, and it begins with a fairly lengthy unskippable cinematic/flyby, then some unskippable tutorial text, then some further unskippable tutorial text on the way in to the castle. There's a lot about it that's not good.


What’s that, two minutes max?

 

I’m currently ten hours into Yakuza 7 and I think about eight hours of that has been cutscenes. 

Link to post
Share on other sites
1 hour ago, jonny_rat said:

I'm sorry to be this person, but I've just started Mario 64 on the all stars collection, and it begins with a fairly lengthy unskippable cinematic/flyby, then some unskippable tutorial text, then some further unskippable tutorial text on the way in to the castle. There's a lot about it that's not good.

 

Is this a wind up? :lol:

 

Unless they've changed it for All-stars, it's an in-engine cutscene that was magical for its time, then two panels of text with about 30 words on each that you skip past by pressing A... then you're in control of Mario. That's all in the space of 1 minute. The later text you get from Lakitu is 6 panels and yep you can keep pressing A straight away if you don't want to read a few sentences - it's over in 10 seconds. 

Link to post
Share on other sites
7 hours ago, SeanR said:

 

ah come on, at the time it was a new controller and they were showing off their new 3D world...

This is partly why SM64 gets away with dropping you in and letting you explore. It relies on the player wanting to tinker with the new controls. A modern game wouldn't get away with putting you in a space like that with very little to do except climb some trees. Nintendo of today would add in some activities that got you used to the controls.

 

6 hours ago, Marlowe said:

 

Is this a wind up? :lol:

 

Unless they've changed it for All-stars, it's an in-engine cutscene that was magical for its time, then two panels of text with about 30 words on each that you skip past by pressing A... then you're in control of Mario. That's all in the space of 1 minute. The later text you get from Lakitu is 6 panels and yep you can keep pressing A straight away if you don't want to read a few sentences - it's over in 10 seconds. 

I think you've missed the point a bit here: length of the opening aside, SM64 is typical of the boring cinematic stuff and clumsy tutorials that teach you in the least natural way possible that you originally mentioned!

 

But here's the thing: in further play, it bakes in teaching the player over the first few levels and star challenges. The first world reinforces the basics and mostly does it without tutorial text. It's the principle of guiding the player through design choices that culminated in all that clever stuff they did in BOTW.

 

7 hours ago, Garwoofoo said:


What’s that, two minutes max?

 

I’m currently ten hours into Yakuza 7 and I think about eight hours of that has been cutscenes. 

 

This isn't just about hating on cutscenes: Y7 is a different beast, and not something i think I could ever get through. This is more about tutorials and ways of teaching players vs letting them review. SM64 might only take 2 minutes from you but it doesn't use that 2 minutes well, and that's fair enough for an old game.

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. Use of this website is subject to our Privacy Policy, Terms of Use, and Guidelines.